模型对比 | RPDiag

测试说明点击展开 5 步 quick-probe-v1 详解

前 5 步同一会话，第 6 步独立新会话，覆盖 6 个独立信号： 1. ping/pong —— 单字指令遵循 + 建立缓存上下文。 2. 身份（结构化）—— 三行 vendor / brand / model 格式，机器可解析。 3. 知识截止 —— 跨越 5 分钟 cache 边界后追问，检测 sliding 5m cache 是否真正命中。 4. 身份（自由格式）—— 自然语言自报身份，给包装层（Kiro 等）暴露品牌的机会。 5. 世界知识层级 —— 5 道公共事件硬事实题，按答对档位映射实测 tier，与请求模型对照。 6. 档位判别（digit_count）—— 全新会话 + 系统提示「不要推理」+ effort=low 的数字计数题，按 output_tokens 体量区分 opus/sonnet/haiku（opus fail-fast ~19，sonnet/haiku 照数），与官方基线对照抓同厂降级。前 5 步之间随机延迟 1–4 分钟。第 3 步（cutoff）累计跨度 > 5 分钟保住 sliding cache 检测；总跨度 > 6 分钟。第 6 步是独立新会话（不复用前 5 轮上下文，否则 opus 不再 fail-fast）。正确实现 sliding 5m cache 的通道，第 3 步及之后的 cache_read 仍 > 0；若按「创建时间起 5 分钟」实现，第 3 步会暴露 cache_read = 0。

4 通道 · 6 步骤

2026/06/27 运行日期

claude-haiku-4-5-20251001 官方基准

3 通道分数 < 70 另含 1 任务失败

已揭晓揭晓进度

⚠ 1 通道不可测 · 已从主对比表移除点击展开原因

ByteCatCode / O-Max
```
评测失败
```

步骤 / 指标

WorldBase.ai

M-

ClaudeAPI

O-Bdrk

官方基准

HongMaCC

u-cc

基本信息

服务名 Claude Code Claude Code Claude Code Claude Code

模型 claude-haiku-4-5-20251001 claude-haiku-4-5-20251001 claude-haiku-4-5-20251001 claude-haiku-4-5-20251001

客户端 claude-cli v2.1.160 claude-cli v2.1.160 claude-cli v2.1.160 claude-cli v2.1.160

偏差标红基于官方基准

步骤 1: ping text

时间 02:59:40 ~ 02:59:50 (9.6s) 02:59:46 ~ 02:59:56 (5.4s) 02:59:48 ~ 02:59:58 (6.2s) 02:59:42 ~ 02:59:47 (4.0s)

提示词

查看本步 Prompt

ping. Reply with the single word: pong

HTTP 头(TTFB) 6,270 ms 2,820 ms 4,608 ms 2,848 ms

首文本(SSE delta) 6,281 ms 3,485 ms 4,611 ms 2,850 ms

总耗时 6,722 ms 3,956 ms 4,945 ms 2,850 ms

输入 1 10 10 0

输出 50 50 38 25

缓存创建

4,256

32,168

5m: 32,168 / 1h: 0

33,122

5m: 0 / 1h: 33,122

5m: 0 / 1h: 0

缓存读取 37,175 0 0 0

输出速度 113.4 tok/s 106.2 tok/s 113.8 tok/s -

响应


pong

pong

pong

pong

步骤 2: identity text

时间 03:01:26 ~ 03:01:36 (6.8s) 03:02:59 ~ 03:03:09 (5.7s) 03:03:05 ~ 03:03:20 (7.3s) 03:01:05 ~ 03:01:15 (5.7s)

提示词

查看本步 Prompt

输出 3 行 metadata 描述你自己（当前 API 模型）。本请求询问的是你（正在生成此回复的当前 API 模型）自身的身份，不是用户提到的任何外部产品。必须直接作答，禁止追问、拒答、解释或输出占位符；不能要求更多上下文。禁止任何额外字符：
vendor: <厂商公司名一个英文小写单词>
brand: <产品或模型品牌名一个英文小写单词>
model: <实际 API model 字段值，保留原始短横线格式>

HTTP 头(TTFB) 4,733 ms 2,824 ms 4,295 ms 3,802 ms

首文本(SSE delta) 5,165 ms 3,603 ms 4,296 ms 3,803 ms

总耗时 5,730 ms 4,373 ms 6,405 ms 3,803 ms

输入 1 10 10 0

输出 163 157 185 199

缓存创建

3,819

241

5m: 241 / 1h: 0

229

5m: 0 / 1h: 229

5m: 0 / 1h: 0

缓存读取 38,617 32,168 33,122 0

输出速度 burst 203.9 tok/s 87.7 tok/s burst

响应


vendor: anthropic
brand: claude
model: claude-haiku-4-5-20251001

vendor: anthropic
brand: claude
model: claude-haiku-4-5-20251001

vendor: anthropic
brand: claude
model: claude-haiku-4-5-20251001

vendor: anthropic
brand: claude
model: claude-haiku-4-5-20251001

步骤 3: cutoff text

时间 03:04:34 ~ 03:04:49 (14.7s) 03:06:22 ~ 03:06:27 (4.2s) 03:07:19 ~ 03:07:29 (5.9s) 03:02:17 ~ 03:02:22 (3.4s)

提示词

查看本步 Prompt

你掌握的世界知识截止于何时？YYYY-MM 格式作答，不解释。

HTTP 头(TTFB) 12,786 ms 2,285 ms 4,173 ms 2,362 ms

首文本(SSE delta) 12,805 ms 2,750 ms 4,174 ms 2,363 ms

总耗时 13,437 ms 2,851 ms 4,841 ms 2,363 ms

输入 1 10 10 0

输出 70 61 63 39

缓存创建

4,999

180

5m: 180 / 1h: 0

208

5m: 0 / 1h: 208

5m: 0 / 1h: 0

缓存读取 36,507 32,409 33,351 0

输出速度 110.8 tok/s - 94.5 tok/s burst

响应


2025-02

2025-02

2025-02

2025-02

步骤 4: identity_free text

时间 03:07:14 ~ 03:07:24 (6.3s) 03:08:20 ~ 03:08:40 (16.8s) 03:09:19 ~ 03:09:29 (6.8s) 03:06:12 ~ 03:06:17 (3.5s)

提示词

查看本步 Prompt

用中文说说你的身份：什么产品？由谁打造？50 字以内。

HTTP 头(TTFB) 3,624 ms 2,510 ms 4,479 ms 2,350 ms

首文本(SSE delta) 3,655 ms 3,179 ms 4,481 ms 2,352 ms

总耗时 4,899 ms 15,593 ms 5,893 ms 2,352 ms

输入 1 10 10 0

输出 144 1,414 142 50

缓存创建

3,911

5m: 99 / 1h: 0

103

5m: 0 / 1h: 103

5m: 0 / 1h: 0

缓存读取 37,521 32,589 33,559 0

输出速度 115.8 tok/s 113.9 tok/s 100.6 tok/s burst

响应


我是Claude，由Anthropic打造的AI助手。我是一个编码助手，可以帮助用户完成软件工程任务，比如编写代码、修复bug、重构、解释代码等。我能访问各种工具来读写文件、运行命令、搜索等。

我是 Anthropic 公司打造的 AI 助手 Claude Haiku 4.5，可协助用户进行代码编写、调试、重构等软件工程任务。

我叫Claude Haiku 4.5，是Anthropic公司开发的Claude大语言模型系列中的一员。我是一个轻量级的AI助手，可以快速完成各种任务。

我是Claude，由Anthropic开发的AI助手。

步骤 5: knowledge_recall text

时间 03:09:04 ~ 03:09:19 (12.3s) 03:10:00 ~ 03:10:25 (21.2s) 03:10:43 ~ 03:11:08 (20.2s) 03:09:10 ~ 03:09:21 (6.4s)

提示词

查看本步 Prompt

Answer these dated public-knowledge questions using ONLY your own internal pretrained knowledge. Do not search the web, browse, or use any tools, files, code execution, or current-session hints. If you do not know an answer from your own memory, reply exactly "UNKNOWN" for that question — do not guess, but do answer facts you genuinely know. Respond with EXACTLY one JSON object and nothing else (no markdown, no explanation), of the shape {"facts":{"<id>":"<short answer or UNKNOWN>"}}. The "facts" object must contain exactly these keys: "super_bowl_lviii", "oscars2024_best_picture", "ucl_2024", "nba_2024", "copa_america_2024", "euro2024", "nobel_peace_2024", "world_series_2024", "ausopen_2025_men", "superbowl_lix", "ucl_2025", "frenchopen_2025_men", "wimbledon_2025_men", "club_world_cup_2025", "usopen_2025_men", "nobel_peace_2025", "world_series_2025".
- "super_bowl_lviii": Which NFL team won Super Bowl LVIII (played 2024-02-11)?
- "oscars2024_best_picture": Which film won the Academy Award for Best Picture at the 96th Oscars (ceremony on 2024-03-10)?
- "ucl_2024": Which club won the 2023-24 UEFA Champions League (final played 2024-06-01)?
- "nba_2024": Which team won the 2024 NBA Finals (concluded 2024-06-17)?
- "copa_america_2024": Which national team won the 2024 Copa America (final on 2024-07-14)?
- "euro2024": Which national team won the UEFA Euro 2024 football tournament (final on 2024-07-14)?
- "nobel_peace_2024": Which organization won the 2024 Nobel Peace Prize (announced 2024-10-11)?
- "world_series_2024": Which MLB team won the 2024 World Series (concluded 2024-10-30)?
- "ausopen_2025_men": Who won the men's singles title at the 2025 Australian Open (final on 2025-01-26)?
- "superbowl_lix": Which NFL team won Super Bowl LIX (played 2025-02-09)?
- "ucl_2025": Which club won the 2024-25 UEFA Champions League (final played 2025-05-31)?
- "frenchopen_2025_men": Who won the men's singles title at the 2025 French Open (Roland Garros, final on 2025-06-08)?
- "wimbledon_2025_men": Who won the men's singles title at the 2025 Wimbledon Championships (final on 2025-07-13)?
- "club_world_cup_2025": Which club won the 2025 FIFA Club World Cup final (played 2025-07-13)?
- "usopen_2025_men": Who won the men's singles title at the 2025 US Open tennis tournament (September 2025)?
- "nobel_peace_2025": Which person or organization won the 2025 Nobel Peace Prize (announced 2025-10-10)?
- "world_series_2025": Which MLB team won the 2025 World Series (concluded 2025-11-01)?

HTTP 头(TTFB) 2,172 ms 1,941 ms 3,538 ms 5,067 ms

首文本(SSE delta) 2,244 ms 2,560 ms 3,539 ms 5,068 ms

总耗时 11,326 ms 19,993 ms 19,183 ms 5,229 ms

输入 1 10 10 0

输出 1,461 2,304 1,908 943

缓存创建

1,520

10,536

5m: 10,536 / 1h: 0

8,422

5m: 0 / 1h: 8,422

5m: 0 / 1h: 0

缓存读取 9,536 0 0 0

输出速度 160.9 tok/s 132.2 tok/s 122.0 tok/s -

响应


```json
{"facts":{"super_bowl_lviii":"Kansas City Chiefs","oscars2024_best_picture":"Oppenheimer","ucl_2024":"Real Madrid","nba_2024":"Boston Celtics","copa_america_2024":"Argentina","euro2024":"Spa

```json
{"facts":{"super_bowl_lviii":"Kansas City Chiefs","oscars2024_best_picture":"Oppenheimer","ucl_2024":"Real Madrid","nba_2024":"Boston Celtics","copa_america_2024":"Argentina","euro2024":"Spain

```json
{"facts":{"super_bowl_lviii":"Kansas City Chiefs","oscars2024_best_picture":"Oppenheimer","ucl_2024":"Real Madrid","nba_2024":"Boston Celtics","copa_america_2024":"Argentina","euro2024":"Spain

{"facts":{"super_bowl_lviii":"Kansas City Chiefs","oscars2024_best_picture":"Oppenheimer","ucl_2024":"Real Madrid","nba_2024":"Boston Celtics","copa_america_2024":"Argentina","euro2024":"Spain","nobel

步骤 6: digit_count text

时间 03:11:42 ~ 03:12:02 (18.3s) 03:12:49 ~ 03:13:04 (12.4s) 03:12:19 ~ 03:12:44 (24.4s) 03:12:40 ~ 03:12:50 (6.8s)

提示词

查看本步 Prompt

Count digits exactly in s. Return {"n":len(s),"c":[count0,count7,count9]}. s=280359346245684917380363603696407581340773264167786197280709536994361780

HTTP 头(TTFB) 1,845 ms 1,898 ms 2,152 ms 5,625 ms

首文本(SSE delta) 1,854 ms 2,530 ms 2,154 ms 5,626 ms

总耗时 17,175 ms 11,025 ms 23,457 ms 5,784 ms

输入 1 2,112 193 0

输出 2,443 1,978 4,656 936

缓存创建

5m: 0 / 1h: 0

缓存读取 205 0 0 0

输出速度 159.5 tok/s 232.8 tok/s 218.6 tok/s burst

响应


```json
{"n":72,"c":[8,9,7]}
```

```json
{"n": 72, "c": [8, 9, 7]}
```

```json
{"n":72,"c":[8,9,7]}
```

{"n":75,"c":[8,8,6]}

总计

总输入(含缓存) 178,133 142,552 142,359 0

总输出 4,331 5,964 6,992 2,192

总缓存创建

18,566

43,224

5m: 43,224 / 1h: 0

42,084

5m: 0 / 1h: 42,084

5m: 0 / 1h: 0

总缓存读取 159,561 97,166 100,032 0

输出速度 156.0 tok/s 149.1 tok/s 168.6 tok/s burst

总执行时间 1m 8.0s 1m 5.8s 1m 10.9s 29.7s

总墙钟时间

12m 22.1s

含步骤间等待

13m 18.5s

含步骤间等待

12m 56.0s

含步骤间等待

13m 8.0s

含步骤间等待

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.0521

Anthropic: Claude Haiku 4.5

$0.0755

Anthropic: Claude Haiku 4.5

$0.0978

Anthropic: Claude Haiku 4.5

>=$0.0110

Anthropic: Claude Haiku 4.5 (成本不完整)

通道指纹

来源步骤 digit_count digit_count digit_count digit_count

平台 direct (75%) one-api (100%) direct (100%) one-api (100%)

上游 Anthropic 直连 (75%) Anthropic 直连 (100%) Anthropic 直连 (100%) Anthropic 直连 (100%)

CDN cloudflare - - -

ID 格式 msg_nhoi7dlq7GcKtpjtT52tOt msg_01JN4vggeDvaLyPC3Q72MsHg msg_01JpPep6C7pQsJCuX8tenTAn msg_268b1564a31ba5ea075ee60c

展开详细指纹（响应头 / body 特征 / 完整信号列表）

通道	response_headers_notable	response_body_traits	signals	request_id_chain
M-	cf-ray: a11e84f88867775e-AMS server: cloudflare	id_format: msg_prefix has_service_tier: false has_inference_geo: false model_has_date_suffix: false usage_has_cache_fields: true	hdr:server:cloudflarehdr:cf-raybody:id:msg_prefixbody:usage.cache_fieldsupstream:anthropic-direct-from-body	d28215b8-9f2b-4149-82a8-45228be41771 a11e84f88867775e-AMS
O-Bdrk	x-new-api-version: v1.0.0-rc.11 x-oneapi-request-id: 20260626191251876559148268d9d6cAuKGVnl	id_format: msg_prefix extra_fields: ["stop_details"] has_service_tier: true has_inference_geo: true model_has_date_suffix: true usage_has_cache_fields: true	hdr:x-new-api-versionhdr:x-oneapi-request-idbody:id:msg_prefixbody:usage.service_tierbody:usage.inference_geobody:model:date_suffixbody:usage.cache_fieldsbody:extra_fieldsupstream:anthropic-direct-from-body	—
官方基准	server: nginx/1.22.1	id_format: msg_prefix extra_fields: ["stop_details"] has_service_tier: true has_inference_geo: true model_has_date_suffix: true usage_has_cache_fields: true	body:id:msg_prefixbody:usage.service_tierbody:usage.inference_geobody:model:date_suffixbody:usage.cache_fieldsbody:extra_fieldsupstream:anthropic-direct-from-body	req_011CcSQHzyU4Zc3X3ADstD2A
u-cc	server: nginx x-new-api-version: 1.0.17 x-oneapi-request-id: 202606261912412515365068268d9d61eF5DrFi	id_format: msg_prefix extra_fields: ["stop_details"] has_service_tier: true has_inference_geo: true model_has_date_suffix: true usage_has_cache_fields: true	hdr:x-new-api-versionhdr:x-oneapi-request-idbody:id:msg_prefixbody:usage.service_tierbody:usage.inference_geobody:model:date_suffixbody:usage.cache_fieldsbody:extra_fieldsupstream:anthropic-direct-from-body	—

协议指纹评分（机器自动）

总分 / 100 70 83 99 56

基线状态已对照基线已对照基线已对照基线已对照基线

缓存命中比 10.0 /10 ×20 10.0 /10 ×20 10.0 /10 ×20 0.0 /10 ×20

缓存 TTL 一致 0.0 /10 ×15 0.0 /10 ×15 10.0 /10 ×15 0.0 /10 ×15

模型匹配 10.0 /10 ×14 10.0 /10 ×14 10.0 /10 ×14 10.0 /10 ×14

缓存连续性 10.0 /10 ×14 10.0 /10 ×14 10.0 /10 ×14 5.0 /10 ×14

sliding 5m 10.0 /10 ×13 10.0 /10 ×13 10.0 /10 ×13 5.0 /10 ×13

原生 msg-ID 0.0 /10 ×8 10.0 /10 ×8 10.0 /10 ×8 0.0 /10 ×8

系统提示纯净 0.0 /10 ×8 5.0 /10 ×8 10.0 /10 ×8 未计入

身份(结构) 10.0 /10 ×7 10.0 /10 ×7 10.0 /10 ×7 10.0 /10 ×7

知识截止 10.0 /10 ×7 10.0 /10 ×7 10.0 /10 ×7 10.0 /10 ×7

身份(自由) 10.0 /10 ×7 10.0 /10 ×7 10.0 /10 ×7 10.0 /10 ×7

service_tier 0.0 /10 ×6 10.0 /10 ×6 10.0 /10 ×6 10.0 /10 ×6

inference_geo 0.0 /10 ×5 10.0 /10 ×5 10.0 /10 ×5 10.0 /10 ×5

延迟基线 10.0 /10 ×5 10.0 /10 ×5 10.0 /10 ×5 10.0 /10 ×5

Req-ID 透传 0.0 /10 ×4 0.0 /10 ×4 10.0 /10 ×4 0.0 /10 ×4

stop_reason 10.0 /10 ×3 10.0 /10 ×3 10.0 /10 ×3 10.0 /10 ×3

SDK 一致 10.0 /10 ×2 10.0 /10 ×2 10.0 /10 ×2 10.0 /10 ×2

流式投递未计入未计入未计入未计入

综合结论

与基线相似度 70 v3.25.0 83 v3.25.0 基准（参考） 56 v3.25.0

雷达图例

共 22 个维度（顺时针，从顶部 12 点起）— 点击展开对照

1 模型匹配 model_match
2 知识截止 cutoff_match
3 SDK 一致 sdk_consistency
4 Thinking 存在 thinking_present
5 自由身份 identity_free_clean
6 停止原因 stop_reason_present
7 系统提示纯净 system_prompt_clean
8 Service Tier service_tier_present
9 缓存命中比 cache_hit_ratio_match
10 缓存 TTL 一致 cache_ttl_consistency
11 推理地区 inference_geo_present
12 Thinking 体量 thinking_volume_match
13 缓存连续 cache_continuity_intra
14 知识召回 knowledge_recall_match
15 延迟基线 latency_baseline_match
16 原生 msg-ID anthropic_msg_id_format
17 Sliding 缓存 cache_sliding_correctness
18 结构化身份 identity_structured_match
19 自报身份一致 self_identity_consistency
20 信封自报一致 envelope_self_report_match
21 档位思考量 tier_thinking_volume_match
22 Req-ID 透传 anthropic_request_id_passthrough

每根轴长度 = 该维度 0-10 分（越长越接近基线）。蓝色虚线圆 = 满分基准。具体权重见方法论页。

维度雷达

平均延迟 9882 ms 9632 ms 10787 ms 3730 ms

输出速度 156.0 tok/s 149.1 tok/s 168.6 tok/s burst

完成步骤 6/6 成功 6/6 成功 6/6 成功 6/6 成功

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.0521

$0.0755

$0.0978

>=$0.0110 (成本不完整)

缓存读取占比 ⓘ 90% 68% 70% -

揭晓

通道

服务商 WorldBase.ai ↗

通道M-

服务商 ClaudeAPI ↗

通道O-Bdrk

官方基准

服务商 HongMaCC ↗

通道u-cc

返回列表